Screening for antimicrobial resistance in fecal samples by the replica plating method.

Abstract

Replica plating can be used for the detection of antibiotic resistance in normal flora. We have evaluated this application of the replica plating method by comparing it with a five-colony method. The replica plating method uses a single plate for each antibiotic, with a concentration just above that for borderline resistance. By the five-colony method, five colonies per sample were picked, chosen to represent all different colony morphologies present, and MICs were determined by a standard agar dilution method. The gram-negative, aerobic floras of 131 fecal samples were screened for resistance to ampicillin, cefuroxime, nalidixic acid, trimethoprim, sulfamethoxazole, and tetracycline by both methods. The rate of resistance detection by the two methods did not differ statistically for any of the antibiotics tested. The breakpoint concentrations used for the replica plates in the study gave results similar to those produced by the agar dilution method and the breakpoint values of the National Committee for Clinical Laboratory Standards and can thus be recommended. As the only currently used resistance detection method, replica plating facilitates an exact determination of the percentage of resistant colonies/total number of colonies (between 1 and 100%) in a sample. This revealed an uneven distribution, with only 23% of the samples having resistance frequencies in the range of 10 to 85%; usually, the resistant flora either was a small minority or was very dominant in samples with resistance. This phenomenon was present for all of the antibiotics.